Small-Dollar Mortgages for Single- Family Residential Properties

HOUSING FINANCE POLICY CENTER RESEARCH REPORT Small-Dollar Mortgages for Single- Family Residential Properties Alanna McCargo Bing Bai Taz George Sarah Strochak URBAN INSTITUTE URBAN INSTITUTE FEDERAL RESERVE BANK OF CHICAGO April 2018 URBAN INSTITUTE

ABOUT THE URBAN INSTITUTE The nonprofit Urban Institute is a leading research organization dedicated to developing evidence-based insights that improve people s lives and strengthen communities. For 50 years, Urban has been the trusted source for rigorous analysis of complex social and economic issues; strategic advice to policymakers, philanthropists, and practitioners; and new, promising ideas that expand opportunities for all. Our work inspires effective decisions that advance fairness and enhance the well-being of people and places. ABOUT THE FEDERAL RESERVE BANK OF CHICAGO The fundamental mission of the Federal Reserve System is to foster the stability, integrity, and efficiency of the nation s monetary, financial, and payment systems in order to promote optimal macroeconomic performance. To this end, the Federal Reserve Bank of Chicago participates in the formulation and implementation of national monetary policy; supervises and regulates state member banks, bank holding companies, and foreign bank branches; and provides financial services to depository institutions and the US government. The Bank s Community Development and Policy Studies division works to improve the socioeconomic prospects of low- and moderateincome people by conducting research and working with community leaders to bring development and reinvestment opportunities to underserved areas. Copyright April 2018. Urban Institute. Permission is granted for reproduction of this file, with attribution to the Urban Institute. Cover image by Tim Meko.

Contents Acknowledgments Executive Summary iv v Small-Dollar Mortgages for Single-Family Residential Properties 1 Defining the Small-Dollar Mortgage Problem 1 State of the Market 1 Why We Need Small-Dollar Mortgages 6 Who Received Small-Dollar Mortgages and through Which Channels? 16 How Small-Dollar Mortgages Can Help in the Home Improvement and Refinance Markets 21 Issues Contributing to the Lack of Small-Dollar Mortgages 23 Recommendations 25 Conclusion 27 Notes 28 References 30 About the Authors 31 Statement of Independence 33

Acknowledgments The Housing Finance Policy Center (HFPC) was launched with generous support at the leadership level from the Citi Foundation and John D. and Catherine T. MacArthur Foundation. Additional support was provided by the Ford Foundation and the Open Society Foundations. Ongoing support for HFPC is also provided by the Housing Finance Innovation Forum, a group of organizations and individuals that support high-quality independent research that informs evidencebased policy development. Funds raised through the Forum provide flexible resources, allowing HFPC to anticipate and respond to emerging policy issues with timely analysis. This funding supports HFPC s research, outreach and engagement, and general operating activities. This report was funded by these combined sources. We are grateful to them and to all our funders, who make it possible for Urban to advance its mission. The views expressed are those of the authors and should not be attributed to the Federal Reserve Bank of Chicago, the Federal Reserve System, the Urban Institute, its trustees, or its funders. Funders do not determine research findings or the insights and recommendations of Urban experts. Further information on the Urban Institute s funding principles is available at urban.org/fundingprinciples. We thank Ellen Seidman and Laurie Goodman from the Urban Institute and Alicia Williams, Mike Berry, Robin Newberger, Susan Longworth, and Gene Amromin from the Federal Reserve Bank of Chicago for reviewing earlier versions of this report. IV ACKNOWLEDGMENTS

Executive Summary This report examines the availability of small-dollar mortgages (up to $70,000) for home purchases, refinances, and improvements. 1 We find evidence of a substantial number of low-cost property sales taking place across many diverse housing markets, but access to credit via traditional mortgage lending is limited for these properties. Low-cost properties could be a larger source of affordable housing if credit access for purchasing and rehabilitating these properties were expanded and improved. The objectives for this report are to define the small-dollar mortgage problem, identify and describe the characteristics of the counties and populations most affected by the lack of small-dollar mortgage options, describe the loan characteristics and production channels of available small-dollar mortgages, and discuss the challenges contributing to the problem and potential solutions for policymakers to consider. The Importance of Low-Cost Properties Homeownership is an important wealth-building mechanism for many American families (Goodman and Mayer 2018). Many first-time homebuyers and low- and middle-income (LMI) families rely on low-cost properties to move from renting to owning a home. Yet, low-cost properties remain largely inaccessible to LMI households because traditional mortgage financing is too difficult to obtain on these properties. 2 The US housing finance system has long failed to meet the needs of people at low socioeconomic levels (Sarkar and McKee 2004), and the problems in low-cost housing markets are one manifestation of this failure. Despite expansion in mortgage finance products for LMI households from the conventional finance market and requirements for banks to lend fairly throughout their assessment areas under the Community Reinvestment Act, LMI borrowers often cannot access available products. Moreover, successful innovations have not been brought to scale (Engel, Keller, and George 2016). Challenges to accessing mortgage credit, including for LMI households, stemming from regulatory changes and tight mortgage lending standards after the latest recession have generated research and discussion (Goodman, Zhu, and George 2014). But the dearth of mortgage credit available for low-cost home EXECUTIVE SUMMARY V

purchases has received less focus. Addressing the small-dollar mortgage problem would open pathways to homeownership for many traditionally underserved groups, including LMI households and people of color, 3 especially in low-cost geographies. This issue disproportionately affects economically challenged communities that have low-cost housing inventory, but the problem is not limited to those areas. Low-cost single-family properties can be found in virtually all cities and in rural areas, and while some of the housing stock is old and dilapidated, there are significant pockets of inventory and sales that constitute decent affordable housing in areas seeing home price appreciation. Factors contributing to the lack of mortgage lending activity for low-cost properties vary across markets and property types. These contextual factors include the severity of the foreclosure crisis and rate of distressed sales, the composition and condition of the residential housing stock, investor competition, and the property appraisal gap. And regardless of local market conditions, mortgage lending standards have been elevated in the postrecession period amid higher origination costs and heightened regulatory scrutiny. Moreover, lenders have less economic incentive to make small loans because they generate lower sales commissions, spreads, and servicing income. Our analysis focuses on traditional purchase financing, but access to financing for home improvement and refinancing for low-cost properties is also limited. Property and structure type also play an important role in determining the financing options, or lack thereof, for low-cost properties. Much manufactured housing is financed as personal property (chattel), with higher rates, shorter terms, and fewer consumer protections than secured mortgage lending. In 2016, only 17 percent of new manufactured homes were titled as real property. 4 Improving access to small-dollar mortgages or introducing new products for the low-dollar segment could support the market for purchasing low-cost single-family and manufactured houses and help LMI borrowers access safe and fairly priced financing for home improvements, repairs, rehabilitation, or refinancing of small-dollar loans. VI EXECUTIVE SUMMARY

Small-Dollar Mortgages for Single- Family Residential Properties Defining the Small-Dollar Mortgage Problem Small-dollar credit can help low-income households meet basic financial needs. Often referred to as microloans, these loans for standard consumer use are typically under $5,000 5 and are personal unsecured loans that have short repayment terms. Microloans for underserved small-business lending communities have gained traction, becoming a source of financing that enables small-business owners to expand and thrive. The concept of small-dollar credit, or microfinancing, for meeting housing needs for low- and moderate-income (LMI) families is still in its infancy in the United States. Extending smalldollar financing, collateralized by real property, may present an opportunity for the underserved lowcost segments of the housing market. For the analysis and recommendations in this report, we define small-dollar mortgages as secured financing for single-family residential properties that, beginning in 2009, are valued between $10,000 and $70,000. 6 Of the single-family homes sold in 2015 in the US, 14 percent, or 643,000 homes, sold for $70,000 or less, of which slightly more than one-fourth were financed with a traditional mortgage loan product. 7 In contrast, among homes worth between $70,000 and $150,000, close to 80 percent of homes sold were financed with a traditional mortgage product. The large proportion of low-cost homes sold without a mortgage reveals the difficulty many prospective homebuyers face in competing against investors and all-cash buyers to purchase affordable homes. We explore these dynamics at the local and national levels, shedding light on the lack of financing and how it affects families and communities. State of the Market The Lack of Small-Dollar Mortgages for Home Purchases Relative to the years before the housing bubble, access to mortgage credit for purchase remains limited and lending standards remain unusually strict, especially borrower credit scores. 8 One result is that 16 percent fewer mortgages were originated in 2016 than in 2001 (when the population was smaller). 9 Although the number of new mortgages has rebounded from the trough in 2011, the growth in lending volume has not extended to mortgages with an origination balance of $70,000 or less. Figure 1, using

Home Mortgage Disclosure Act (HMDA) data, shows the number of first-lien, owner-occupied, singlefamily purchase originations from 2009 to 2016 by origination amount. In 2009, there were 205,797 new mortgages with a balance between $10,000 and $70,000, or 8 percent of all originations. By 2016, the number of loans in this range had fallen 17 percent and made up only 5 percent of all originations. Originations between $70,000 and $150,000 declined less than 1 percent. Loans between $150,000 and $300,000 increased 61 percent, and loans greater than $300,000 increased 142 percent, nearly doubling in market share. Home price appreciation accounts for some of the shift. From January 2009 to December 2016, the national Home Price Index increased 33 percent while the market share of new loans up to $70,000 decreased 43 percent. FIGURE 1 New Mortgages by Origination Amount $10,000 $70,000 $70,000 $150,000 $150,000 $300,000 > $300,000 1,800,000 1,600,000 1,400,000 1,200,000 1,000,000 800,000 600,000 400,000 200,000 0 2009 2010 2011 2012 2013 2014 2015 2016 Source: Home Mortgage Disclosure Act records. Note: Includes first-lien, owner-occupied, single-family purchase originations. URBAN INSTITUTE Only One in Four Homes Sold for $70,000 or Less Has a Mortgage We can see the scarcity of small-dollar home loans by comparing the share of purchase mortgages originated for $70,000 or less with the share of all home sales for $70,000 or less. 10 Nationwide, the share of mortgages and small-dollar home sales have both declined, especially when compared with the 2 SMALL- DOLLAR MORTGAGES FOR SINGLE- FAMILY RESIDENTIAL PROPERTIES

home price rebound after the crisis (figure 2). In 2015, only 5.5 percent of purchase mortgages were $70,000 or less, about 40 percent of the low-cost home sale share of 13.9 percent. This mismatch of small mortgages is even more apparent when we compare the number of new mortgages of $70,000 or less with the number of homes sold for $70,000 or less. In 2015, more than 643,000 homes sold for $70,000 or less, yet just over 177,000 borrowers took out a mortgage of $70,000 or less to purchase a home, accounting for 27.5 percent of home sales (figure 3). In other words, for homes sold for $70,000 or less, one in four sales was financed with a mortgage. The share of small-dollar home sales financed by mortgages has been between 25 and 29 percent from 2010 to 2015, highlighting the fact that this is not a new problem and that financing has been constrained for this part of the market for some time. FIGURE 2 Share of Home Purchases and Home Sales up to $70,000 25% 20% 15% Share of mortgages up to $70,000 Share of home sales up to $70,000 Home price index 21.7% 19.8% 19.7% 18.6% 17.1% 15.6% 13.9% 200 180 160 140 120 100 10% 8.4% 8.6% 9.2% 8.1% 7.0% 6.4% 5.5% 80 60 5% 40 20 0% 2009 2010 2011 2012 2013 2014 2015 0 URBAN INSTITUTE Sources: Home Mortgage Disclosure Act records and CoreLogic Home Price Index. Note: Includes first-lien, owner-occupied, single-family purchase originations. SMALL- DOLLAR MORTGAGES FOR SINGLE- FAMILY RESIDENTIAL PROPERTIES 3

FIGURE 3 Share of Sales Financed by Mortgages 87.9% Home sales $70,000 Home sales > $70,000 80.7% 77.7% 79.3% 75.3% 72.6% 73.5% 32.9% 29.1% 26.4% 25.6% 26.1% 27.6% 27.5% 2009 2010 2011 2012 2013 2014 2015 Sources: Home Mortgage Disclosure Act records and CoreLogic. URBAN INSTITUTE Comparison of Sales and Mortgages by Price Buckets We compare these numbers with homes purchased in the $70,000-to-$150,000 price range. Home sales in this higher price range are more likely to be financed by a traditional mortgage. In 2015, 79.3 percent of home sales between $70,000 and $150,000 were mortgage financed (figure 3). The difference between the mortgage shares of low-cost homes and more expensive homes has persisted, despite the recent housing market recovery, indicating continued strain on accessing credit for low-cost properties. Home purchases that do not involve a mortgage are generally paid for with cash and are often distressed sales or foreclosures sold through auction or other distressed-asset programs. Without access to mortgage products for small dollar amounts, many creditworthy LMI households that could afford a home cannot compete to purchase these properties. Why We Define Small-Dollar Mortgages at $70,000 In prior work, we defined small mortgage loans as having dollar amounts up to $50,000. For this analysis, we propose a new dollar amount, $70,000, based on an observed decline in mortgage financing 4 SMALL- DOLLAR MORTGAGES FOR SINGLE- FAMILY RESIDENTIAL PROPERTIES

relative to sales volume at this level. 11 Figure 4 represents the share of sales financed by a mortgage in various price buckets. A typical home purchase transaction usually involves a down payment. For context, a $70,000 mortgage could be for a $72,450 home with a 3.5 percent down payment up to an $84,000 home with a 20 percent down payment. This price range for home purchase is affordable in many metropolitan and rural areas. This definition of small-dollar mortgages is intended to show what price point potential borrowers are likely to encounter difficulty in obtaining a traditional mortgage loan via standard mortgage channels. Having a standard definition of small-dollar mortgages will also help policymakers and industry stakeholders define solutions and alternatives for the lack of housing finance options in this price range. In 2015, 8 percent of properties sold between $10,000 and $30,000 were financed by a mortgage (figure 4). The share is 48 percent for properties sold between $50,000 and $70,000, still less than half of total home sales in that range. The mortgage-financed share jumps up to 70 percent for homes sold between $70,000 and $100,000. Because of the financing disparity between these price buckets, we established the limit for small-dollar mortgages at $70,000. Although we use $70,000 as a threshold, many factors affecting small-dollar mortgages are relevant to home sales above the threshold, especially in high-cost markets where a sale of $125,000, for example, would be considered small. FIGURE 4 Share of Sales Financed by Mortgages 100% $10,000 $30,000 $30,000 $50,000 $50,000 $70,000 $70,000 $100,000 $100,000 $150,000 $150,000 $300,000 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 2009 2010 2011 2012 2013 2014 2015 Sources: Home Mortgage Disclosure Act records and CoreLogic. URBAN INSTITUTE SMALL- DOLLAR MORTGAGES FOR SINGLE- FAMILY RESIDENTIAL PROPERTIES 5

How Does Recent Price Appreciation Affect Small-Dollar Mortgages? Home prices bottomed out in 2011 and increased 33 percent through 2016. From 2011 to 2016, the share of owner-occupied homes valued at $70,000 or less dropped from 14 percent to 12 percent (see figure 11, page 13), the share of sales at $70,000 or less fell from 22 percent to 14 percent, and the share of purchase loans of $70,000 or less decreased from 9 percent to 6 percent (see figure 2, page 3). We set our small-dollar mortgage threshold at $70,000 for all years using nominal dollars. The data on share of sales financed by a mortgage (see figure 4, previous page) suggest that despite the change in home prices, the share of home sales financed by mortgages for each price bucket stayed stable. That is, as higher prices pushed more homes into higher price buckets, the share of the sales at $70,000 or less purchased with mortgages still hovered between 25 and 28 percent (see figure 3, page 4). An alternative way to set the small-dollar mortgage limit would be to adjust it with home prices over time. The share of sales at $70,000 or less fell from 22 percent to 14 percent from 2011 to 2015 because of home price appreciation. If we set the limit at $70,000 in 2011 and adjust it to $100,000 in 2015, sales from $70,000 to $100,000 would add another 8 percent of sales in 2015, making the smalldollar-sales share with the higher limit in 2015 about the same level as 2011 at 22 percent. However, because sales between $70,000 and $100,000 have had a much higher mortgage share (figure 4), the share of small-dollar home sales financed would be 43 percent in 2015, much higher than 2011 s 26 percent with the lower $70,000 small mortgage limit. Instead, we impose a constant nominal $70,000 cutoff, which results in a stable mortgage sales share from 2011 to 2015 for the sales up to the same limit in figure 3. Why We Need Small-Dollar Mortgages Small-Dollar Home Sales Are Widespread throughout the US Housing affordability and the challenges of high-cost markets such as San Francisco get much attention in the press and academic research, but many communities have a significant share of low-cost home sales. Figure 5 shows the share of home sales up to $70,000 for every US county with data available. 12 Suburban, rural, and urban counties have sales in this price range, and in many counties, small-dollar sales make up most home sales. If creditworthy potential homebuyers cannot obtain financing for these houses, they may miss out on that important first rung on the homeownership ladder that helps families and neighborhoods gain economic stability and begin wealth building through ownership. 6 SMALL- DOLLAR MORTGAGES FOR SINGLE- FAMILY RESIDENTIAL PROPERTIES

FIGURE 5 Share of Home Sales up to $70,000, by County URBAN INSTITUTE Sources: CoreLogic and the US Census Bureau. Where Are These Low-Cost Counties? To better understand the housing markets that have large concentrations of low-cost properties, we selected the 300 counties with the highest shares of home sales up to $70,000 among all counties that had at least 500 home sales in 2015. These low-cost counties are highlighted in figure 6. There were 449,000 homes sold in the low-cost counties in 2015, accounting for about 10 percent of all US home sales. Thirty-nine percent of these home sales were for $70,000 or less, compared with only 14 percent in the US (figure 7). The high concentration of homes in the low-cost counties means that the lack of mortgage lending available in these markets affects a larger share of the housing market. Twenty-one percent of low-cost counties are not part of a core-based statistical area, 13 while 42 percent are part of a micropolitan statistical area and the remaining 37 percent are part of a SMALL- DOLLAR MORTGAGES FOR SINGLE- FAMILY RESIDENTIAL PROPERTIES 7

metropolitan statistical area. The counties include a mix of almost or entirely rural areas, counties that include small cities, and counties that are part of large metropolitan areas. They include counties that contain larger cities, such as Wayne County, Michigan (Detroit), and Cuyahoga County, Ohio (Cleveland), and others that are predominantly suburban outlying areas of large cities, such as Clayton County, Georgia (adjacent to Atlanta). The actual and potential strength of the housing markets in these counties depend not only on mortgage credit access but on broader considerations, such as demographic trends, economic activity, and the quality and proximity of schools, jobs, health care facilities, and basic services. In many cases, these markets overlap with distressed areas that have struggled to recover from the foreclosure crisis and from long-standing trends of low (or declining) population growth. Other low-cost counties include rural areas where housing is generally less costly and where manufactured housing sometimes makes up a substantial market share. As with metropolitan low-cost counties, rural low-cost counties cannot be easily generalized. For example, nearly all of Oklahoma s 23 low-cost counties overlap with tribal jurisdiction areas. FIGURE 6 300 Counties with the Highest Shares of Home Sales up to $70,000 Sources: CoreLogic and the US Census Bureau. URBAN INSTITUTE 8 SMALL- DOLLAR MORTGAGES FOR SINGLE- FAMILY RESIDENTIAL PROPERTIES

FIGURE 7 Share of Home Sales up to $70,000 44% 45% Low-cost counties 48% 47% 44% US 42% 39% 19% 20% 22% 20% 17% 16% 14% 2009 2010 2011 2012 2013 2014 2015 URBAN INSTITUTE Sources: CoreLogic and the US Census Bureau. Who Lives in Low-Cost Counties? Households in low-cost counties are more likely to be low income and live in older houses in need of repairs (table 1). Although low home prices may suggest it is more economical to own a home in low-cost areas, homeownership rates are only slightly higher than the US average, at least in part because of the lack of small-dollar mortgage financing. In 2015, only 21 percent of homes that sold for $70,000 or less in low-cost counties were financed by a traditional mortgage (figure 8). TABLE 1 Demographic Comparison of Low-Cost Counties and Other US Counties Median Income ($) Minority Share (%) Houses Built before 1980 (%) Homeownership Rate (%) Low-cost counties United States Low-cost counties United States Low-cost counties United States Low-cost counties United States 2009 41,000 50,800 20 24 51 42 69 66 2010 40,200 50,000 20 25 49 41 69 65 2011 40,400 50,000 20 25 48 41 69 64 2012 41,600 51,500 21 25 49 40 68 64 2013 42,500 52,700 21 25 48 40 67 63 2014 43,500 54,200 21 26 48 40 67 63 2015 45,000 56,400 21 26 47 39 67 63 2016 46,000 58,200 22 26 47 39 67 63 Source: American Community Survey. SMALL- DOLLAR MORTGAGES FOR SINGLE- FAMILY RESIDENTIAL PROPERTIES 9

FIGURE 8 Share of Sales Financed by a Mortgage in Low Cost Counties 76.3% Home sales $70,000 Home sales $70,000 $150,000 71.1% 69.7% 67.1% 67.2% 67.7% 69.5% 27.3% 24.5% 22.6% 21.3% 20.8% 21.2% 21.3% 2009 2010 2011 2012 2013 2014 2015 Sources: Home Mortgage Disclosure Act records and CoreLogic. URBAN INSTITUTE We can gain insight about who would be served by small-dollar mortgages by looking at the household demographics of renters in low-cost markets and owners of low-cost homes. Demographic characteristics of households in homes valued up to $70,000 and households currently renting are good indicators of potential borrowers who could benefit from small-dollar mortgages and move from renting to owning. Figure 9 summarizes the demographic insights for the US, and figure 10 provides similar information for low-cost counties. 10 SMALL- DOLLAR MORTGAGES FOR SINGLE- FAMILY RESIDENTIAL PROPERTIES

FIGURE 9 Demographic Characteristics of US Renters and Homeowners, 2016 37% $80,000 21% $55,000 16% $38,000 $37,000 7% Share of households Median household income 39.0% 50% 46% 27% 40% 20% 30% 16% Minority share Share built before 1980 Source: American Community Survey. Nationwide, 43 million households were renters in 2016, or 37 percent of all households (figure 9). The median income of renters is comparable with that of homeowners who live in homes valued up to $70,000, indicating that some renters could move into homeownership if they had access to small-dollar mortgages. Such access could also help minority and low-income households become homeowners. A higher share of renters than homeowners are minorities; in 2016, the minority share of renters was 39 percent, higher than the 27 percent of homeowners in homes worth $70,000 or less. SMALL- DOLLAR MORTGAGES FOR SINGLE- FAMILY RESIDENTIAL PROPERTIES 11

FIGURE 10 Demographic Characteristics of Renters and Homeowners in Low-Cost Counties, 2016 $82,200 34% 28% $55,900 16% 17% $29,000 $36,500 Share of households Median household income 35% 62% 22% 49% 53% 12% 29% 8% Minority share Share built before 1980 Source: American Community Survey. Nearly 34 percent of households in low-cost counties, or 3.7 million, are renting (figure 10). These renter households have the lowest median income ($29,000) and highest minority share (35 percent) compared with other households. The median renter income is 20.5 percent below the median income of households who own a house worth $70,000 or less, but the median renter household with income of $28,000 could afford a $127,000 house with zero down payment if the median family can use 34 percent of its income to pay for a 30-year fixed-rate mortgage, at a 4.4 percent mortgage rate, and property tax and insurance at 1.75 percent of the house s value. 14 This suggests that many renters might 12 SMALL- DOLLAR MORTGAGES FOR SINGLE- FAMILY RESIDENTIAL PROPERTIES

be in the market for low-cost starter homes if they had adequate credit scores and could access reasonably priced small-dollar mortgages. 15 Existing Low-Cost Owner-Occupied Housing Stock Compared with the country as a whole, a larger share of the owner-occupied housing stock in low-cost counties is worth $70,000 or less. In 2016, 12 percent of US owner-occupied homes were worth $70,000 or less. In the low-cost counties, this share was 24 percent (figure 11). FIGURE 11 Share of Owner-Occupied Homes Worth $70,000 or Less Low-cost counties US 24.5% 25.2% 26.9% 27.7% 27.6% 26.8% 25.4% 24.3% 13.1% 13.4% 14.1% 14.6% 14.2% 13.5% 12.1% 11.5% 2009 2010 2011 2012 2013 2014 2015 2016 Source: American Community Survey. URBAN INSTITUTE Because of home price appreciation, the availability of low-cost housing has decreased. From 2012 to 2015, the share of the US housing stock worth $70,000 or less decreased from 15 percent to 12 percent. Homeowners who purchased low-cost homes are benefiting from this appreciation, while renters in markets that experienced rapid appreciation may have been displaced and priced out of the market. SMALL- DOLLAR MORTGAGES FOR SINGLE- FAMILY RESIDENTIAL PROPERTIES 13

Small-Dollar Mortgages for Home Renovation Small-dollar mortgages for renovation could help homeowners in low-cost markets maintain their homes, deal with rising costs, or realize the benefits of neighborhood price appreciation. A small-dollar financing could help return distressed rental or vacant property to the owner-occupied housing stock or help landlords of rental properties improve homes for tenants and revitalize their communities. Fifty percent of US homes worth $70,000 or less were built before 1980, and the share was 62 percent in low-cost counties. Low-cost homes are often older than high-priced properties, and many need renovation. The Neighborhood Stabilization Program, 16 a program put in place in 2008 amid the foreclosure crisis, allocated emergency funding to states and localities so they could help purchase, rehabilitate, redevelop, and resell foreclosed or abandoned properties in distressed neighborhoods. As those programs expire or suffer budget cuts, new ideas are emerging that can become a catalyst for expanding access to small-dollar mortgages for home purchase and renovation, supporting communities, and directing needed capital into communities for long-term sustainability. Detroit Land Bank s Rehabbed and Ready program 17 helps buyers buy and renovate homes and put them back to productive use. Another example is Philadelphia s Healthy Rowhouse Project, 18 which is improving substandard conditions in row houses occupied by LMI families. Low-Cost Homes Have More Frequent Home Sales Low-cost homes are sold more frequently than their high-valued counterparts. In 2015, sales of homes worth $70,000 or less across the US accounted for 7.4 percent of the owner-occupied housing stock, compared with a 5.5 percent turnover of homes worth $70,000 to $150,000. This trend is more pronounced in low-cost counties, where there were 182,000 sales of homes worth $70,000 or less in 2015, accounting for nearly 10 percent of the nearly 1.9 million owneroccupied households worth $70,000 or less in these areas. In comparison, only 5 percent of homes worth $70,000 to $150,000 turned over in 2015. 14 SMALL- DOLLAR MORTGAGES FOR SINGLE- FAMILY RESIDENTIAL PROPERTIES

FIGURE 12 Home Sales as a Share of Housing Stock in Low-Cost Counties and the US 7.6% 7.7% 7.7% Low-cost counties $70,000 $70,000 $150,000 8.2% 8.9% 9.1% 9.7% 3.4% 3.2% 3.1% 3.5% 4.2% 4.5% 5.0% 2009 2010 2011 2012 2013 2014 2015 US $70,000 $70,000 $150,000 6.7% 6.9% 7.2% 7.1% 7.2% 6.8% 7.4% 4.9% 4.5% 4.4% 4.8% 5.2% 5.2% 5.5% 2009 2010 2011 2012 2013 2014 2015 Sources: CoreLogic and American Community Survey. URBAN INSTITUTE A reasonable hypothesis for this difference can be partly attributed to non owner occupants (e.g., investors in single-family residential property) purchasing low-cost homes. Just 27.5 percent of US home sales up to $70,000 in 2015 were financed by a mortgage, and the share is 21.3 percent in lowcost counties. Home sales without a mortgage were primarily cash sales, most by investors who fixed up and rented or flipped the homes. Some of the housing was seller financed, and others represented sales managed by land bank authorities. SMALL- DOLLAR MORTGAGES FOR SINGLE- FAMILY RESIDENTIAL PROPERTIES 15

Who Received Small-Dollar Mortgages and through Which Channels? Characteristics of Small-Dollar Mortgage Borrowers Data from HMDA allow us to describe demographic attributes of households that could obtain a smalldollar purchase mortgage. Such borrowers represent only a fraction of small-dollar homebuyers and of households (owners and renters) who live in low-cost homes. Nevertheless, the data reveal how access to small mortgages could make homeownership attainable for more people. INCOME Income is a critical factor in obtaining financing for a home purchase. Lower loan amounts are associated with lower borrower incomes, based on HMDA data from 2009 to 2016. A typical borrower of a mortgage loan up to $70,000 has a median income of about $35,000, compared with a median income of $71,000 among all borrowers. Small-loan borrowers also have lower typical core-based statistical area and census tract level incomes than all borrowers overall, meaning small loans are more likely to occur in low-income core-based statistical areas and neighborhoods. 19 Other key neighborhood differences between small-loan borrowers and large-loan borrowers are that small loans typically have a higher neighborhood vacancy rate (median of 9 percent versus 6 percent) and a lower neighborhood homeownership rate (median of 75 percent versus 79 percent). These figures include only approved, originated loans. Homeownership rates may be lower in markets where potential smallloan buyers remain renters because of a lack of financing available to fit their needs. BORROWER RACE OR ETHNICITY The racial and ethnic composition of small-dollar mortgage borrowers varies modestly from borrowers overall, according to data from HMDA and the American Community Survey. Among borrowers with loan amounts up to $70,000, 76 percent are white, compared with 71 percent of borrowers overall (figure 13). The loan share of black borrowers is slightly higher for small-dollar mortgages than for all loans, while the loan share for Hispanic borrowers is the same as the overall share. The share of loans to Asian borrowers is only 2 percent for small-dollar mortgages but is 6 percent overall. The share of borrowers for whom race or ethnicity information is unavailable increases as loan size increases, making it difficult to interpret, as the borrowers without race or ethnicity information could disproportionately come from one racial or ethnic group. 16 SMALL- DOLLAR MORTGAGES FOR SINGLE- FAMILY RESIDENTIAL PROPERTIES

FIGURE 13 Purchase Loan Originations by Race or Ethnicity 76% 71% Small-dollar mortgages All mortgages 7% 6% 10% 10% 2% 6% 6% 8% White Black Hispanic Asian Not available URBAN INSTITUTE Source: Home Mortgage Disclosure Act records. Notes: Based on purchase mortgage originations from 2009 to 2016. White people and black people are non-hispanic. Loan Type We analyzed HMDA data to shed light on the mortgage financing channels primarily used for smalldollar mortgage originations. The loan type field indicates whether each originated loan type was from the Federal Housing Administration (FHA), the US Department of Veterans Affairs (VA), or the Rural Housing Service or was from conventional financing, with the conventional category including government-sponsored enterprise (GSE) and portfolio loans. We then used loan-level single-family mortgage data from the GSEs to further break down the conventional type into GSE and portfolio loans. Because the GSE data were available only since 2013, we analyzed purchase loans by loan type originated from 2013 to 2016 (figure 14). Government-sponsored enterprise loans compose 53 percent of all originations but only 45 percent of small-dollar mortgage originations. The FHA serves 19 percent of the small-dollar mortgage market and 24 percent of the overall market, while Rural Housing Service loans serve 6 percent of small-dollar mortgages and 4 percent of the overall market. The largest gaps in market share of small loans relative to the overall market are in portfolio and VA lending, though in the opposite direction. The VA financed about 10 percent of purchase loans from 2013 to 2016 but only 3 percent of small-dollar purchase mortgages. Veterans Administration lending for low-cost properties may be particularly affected by the SMALL- DOLLAR MORTGAGES FOR SINGLE- FAMILY RESIDENTIAL PROPERTIES 17

VA s residual income test, which could require buyers at small dollar amounts to have a low debt-toincome ratio to have sufficient residual income to qualify for a loan. Twenty-eight percent of small purchase loans were retained in portfolio, which is more than three times as large as its portion of all purchase loans. This significantly higher share of small mortgages in portfolio is largely because many of these loans are originated and retained by small community banks, credit unions, and large lenders who work with local partners and are sources of liquidity where secondary market options are not as accessible. These small-dollar mortgages can also have unique servicing needs that lenders wish to closely manage. FIGURE 14 Market Share by Loan Type Small-dollar mortgages All mortgages 53% 45% 28% 24% 19% 9% 3% 10% 6% 4% GSE Portfolio FHA VA RHS URBAN INSTITUTE Sources: Home Mortgage Disclosure Act records and embs. Notes: Based on purchase mortgage originations from 2013 to 2016. FHA = Federal Housing Administration; GSE = governmentsponsored enterprise; RHS = Rural Housing Service; VA = US Department of Veterans Affairs. Borrower Credit Characteristics Although HMDA provides rich information on borrower demographics and loan types, it does not contain key credit risk characteristics, such as borrower credit score, loan-to-value ratio, and debt-toincome ratio. HMDA also has a one-year lag, with the 2017 HMDA origination data scheduled to be 18 SMALL- DOLLAR MORTGAGES FOR SINGLE- FAMILY RESIDENTIAL PROPERTIES

released in 2018. To examine the credit profiles of borrowers who obtained small-dollar purchase mortgages, we used complete loan-level single-family purchase-money mortgage data released by the GSEs (i.e., Fannie Mae and Freddie Mac) and by Ginnie Mae, which securitizes government-guaranteed loans from the FHA, VA, and US Department of Agriculture. Most first homes have been financed through these channels in recent years. The data we used for 2016 contain more than 70 percent of new purchase loans. Notably, these data exclude portfolio originations, so originations with a loan amount exceeding the agency limits are not included. Although about 28 percent of small-dollar mortgages were made through the portfolio channel, GSE and government loans cover about 72 percent of the market of loans up to $70,000. Table 2 shows the credit characteristics by loan amount and agency for purchase loans originated in 2017 for small-dollar and larger purchase mortgages. New small-dollar and large mortgage borrowers exhibit high median FICO scores, with small-dollar borrowers scores slightly lower across all channels (722 versus 730). But small-dollar mortgages are also associated with lower loan-to-value and debt-toincome ratios, with the typical small-dollar borrower in the GSE channel making a 20 percent down payment. The interest rate on loans up to $70,000 is higher across all channels, with the widest disparity for VA loans (4.25 percent for loans up to $70,000 compared with 3.88 percent for loans above $70,000). The share of first-time homebuyers is similar for small-dollar and large mortgages across all channels, yet small-dollar mortgage borrowers in the FHA and VA channels are more likely to be firsttime homebuyers than large-loan borrowers in those channels. SMALL- DOLLAR MORTGAGES FOR SINGLE- FAMILY RESIDENTIAL PROPERTIES 19

TABLE 2 Borrower Credit Characteristics by Loan Amount 2017 purchase mortgage originations in the agency market GSE FHA VA Other All $70K >$70K $70K >$70K $70K >$70K $70K >$70K $70K >$70K Loan count 74,417 1,837,146 24,240 779,478 4,111 359,619 7,739 110,699 110,507 3,086,942 Share 4% 96% 3% 97% 1% 99% 7% 93% 3% 97% First timehomebuyer share 40% 40% 83% 76% 69% 53% 86% 81% 53% 53% Owner-occupied share 70% 87% 100% 100% 100% 100% 100% 100% 80% 92% Median loan amount $56,610 $222,562 $59,000 $189,000 $60,000 $239,000 $60,000 $141,000 $58,000 $211,000 Credit score 748 758 667 672 690 711 686 694 722 730 Loan-to-value ratio 80.0% 80.0% 96.5% 96.5% 100.0% 100.0% 101.0% 101.0% 86.8% 95.0% Debt-to-income ratio 33.00% 36.00% 37.02% 43.52% 33.79% 41.02% 32.45% 36.17% 33.58% 38.94% Note rate 4.38% 4.25% 4.38% 4.13% 4.25% 3.88% 4.25% 4.00% 4.38% 4.13% Source: embs. Note: FHA = Federal Housing Administration; GSE = government-sponsored enterprise; VA = US Department of Veterans Affairs. 20 SMALL- DOLLAR MORTGAGES FOR SINGLE- FAMILY RESIDENTIAL PROPERTIES

How Small-Dollar Mortgages Can Help in the Home Improvement and Refinance Markets The lack of small-dollar mortgages also affects home refinance and home improvement lending. But origination costs of certain refinance and home improvement loans are lower when the borrower qualifies for a waiver of appraisal or other fees. For example, the FHA offers a streamlined refinance program for FHA borrowers that requires no appraisal and no credit report. The mismatch between housing stock and small mortgage originations affects refinance and purchase loans (figure 15). In 2016, 11.5 percent of owner-occupied houses were worth $70,000 or less, while only 5.4 percent of refinance mortgages were made for $70,000 or less. The lack of small-dollar refinance mortgages for owners of low-cost properties could be more serious than these numbers indicate because many of these small-dollar refinance loans were taken out by homeowners of more expensive homes who paid down their mortgages to $70,000 or less. FIGURE 15 Mismatch between Small-Dollar Refinance Originations and Low-Cost Housing Stock Share of refinance loans up to $70,000 Share of owner-occupied houses up to $70,000 13.1% 13.4% 14.1% 14.6% 14.2% 10.4% 13.5% 12.1% 11.5% 8.9% 9.3% 6.4% 6.9% 7.5% 6.8% 5.4% 2009 2010 2011 2012 2013 2014 2015 2016 Sources: Home Mortgage Disclosure Act records and CoreLogic. URBAN INSTITUTE SMALL- DOLLAR MORTGAGES FOR SINGLE- FAMILY RESIDENTIAL PROPERTIES 21

A few small-dollar refinance originations continued even during refinance boom years in 2012 and 2015, when refinance originations up to $70,000 did not increase as much as higher-balance loan refinances; small refinance loans even declined in 2015 (table 3). The lack of a small-dollar refinance credit makes it more difficult for owners of low-cost homes, who are more likely to be low-income households, to refinance their mortgages to more favorable rates and terms or leverage home equity. The scarcity of small-dollar mortgage credit extends to home improvement loans. These are especially important to owners of low-cost homes, as these properties are more likely to be older or to have been purchased in a distressed sale and could benefit from a cost-effective renovation to continue or return these homes to safe, habitable, and affordable housing. Figure 16 shows that the share of small-dollar home improvement loans up to $50,000 has been low historically and has declined rapidly. For buyers and owners of low-cost properties, access to small financing amounts for improvements can be a significant opportunity for an LMI household that needs to maintain or rehabilitate a home they own or would like to purchase. Constraining home improvement lending to these owners exacerbates problems with upkeep in these communities. TABLE 3 Refinance Loans by Price Bucket Number of Refinance Loans Share of refinance loans $70,000 $70,000 $150,000 >$150,000 $70,000 2009 332,225 1,610,033 3,341,355 6.3% 2010 308,373 1,434,479 2,763,552 6.8% 2011 341,061 1,281,205 2,215,507 8.9% 2012 440,152 1,816,378 3,642,154 7.5% 2013 403,657 1,505,366 2,473,871 9.2% 2014 206,363 667,972 1,118,719 10.4% 2015 192,534 785,785 1,861,279 6.8% 2016 179,236 831,806 2,358,641 5.3% Source: Home Mortgage Disclosure Act records. 22 SMALL- DOLLAR MORTGAGES FOR SINGLE- FAMILY RESIDENTIAL PROPERTIES

FIGURE 16 Home Improvement Loans by Price Bucket 100% $10,000 $10,000 $30,000 $30,000 $50,000 $50,000 $70,000 $70,000 $150,000 $150,000 $300,000 >$300,000 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 2009 2010 2011 2012 2013 2014 2015 2016 Source: Home Mortgage Disclosure Act records. URBAN INSTITUTE Issues Contributing to the Lack of Small-Dollar Mortgages Many factors contribute to the lack of small-dollar mortgages. Some factors are tied to local market conditions, including severity of the foreclosure crisis and resulting distressed properties, the composition and condition of the residential housing stock, investor competition, and the property appraisal gap. In markets that have experienced population decline and high vacancy rates, an appraisal gap can occur when a home for sale fails to appraise up to the value necessary for a lender to underwrite a loan for the price consistent with what the purchaser has agreed to pay. The appraisal gap is not specific to small-dollar loans, but some areas experiencing this challenge are among those with the greatest concentrations of small-dollar sales, such as Wayne County, Michigan. In many low-cost housing markets, competition from investors purchasing with cash suppresses would-be low-cost property buyers. Sellers eager for the speed and certainty of an all-cash closing give an advantage to investors, who will typically rent out the property, over an owner-occupant purchasing SMALL- DOLLAR MORTGAGES FOR SINGLE- FAMILY RESIDENTIAL PROPERTIES 23

with a mortgage. Cash sales are quicker and more certain because they do not involve a lender and usually have reduced appraisal and inspection requirements. This issue may not be solved by a traditional mortgage product but may require alternative strategies (e.g., community land trusts with funds to act quickly when homes come on the market) to give first-time homebuyers and owneroccupants a fair shot at purchasing low-cost homes. Housing stock composition affects whether mortgage lending is viable for low-cost properties. Manufactured homes can be difficult or impossible to finance with a traditional mortgage, especially when the house will be placed on leased land, often leading prospective buyers to turn to more costly and risky forms of credit. Borrowers who owned land and took out a loan to purchase a manufactured home often financed the home purchase with a chattel loan (CFPB 2014). To a lesser degree, low-cost condominium units can be more complicated to finance with a mortgage, especially via the FHA, but condos can be a source of more affordable ownership in certain markets. Irrespective of local market conditions, many factors contributing to today s tight mortgage credit box are exacerbated for borrowers seeking loans for low-cost properties. Since the Great Recession, lenders have reported higher origination costs, greater regulatory scrutiny on representations and warranties and indemnification issues subjecting them to higher costs of delinquency, higher servicing costs for nonperforming loans, and reputational and monetary penalties from mortgage settlements in the aftermath of the crisis. Mortgage lenders seek to compensate for these higher costs with higher profits on individual loans. But the structure of mortgage lending compensation and incentives works to the disadvantage of smalldollar loans. Loan origination costs are largely fixed and recovered either through the sale of the loan or through the financing spread and payment for servicing. Smaller loans generate lower sales prices, spreads, and servicing income, making them less economically attractive to lenders. What then are the options for transacting on low-cost properties? The three possibilities are no sales, more all-cash purchases (which tend to be to investors, rather than owner-occupants), and greater use of seller financing vehicles, including land contracts, which often have fewer protections for borrowers than traditional mortgages. None of these is optimal for potential homeowners or communities who might benefit from purchasing or renovating low-cost properties for use by owneroccupants. 24 SMALL- DOLLAR MORTGAGES FOR SINGLE- FAMILY RESIDENTIAL PROPERTIES

Recommendations Evidence suggests that it is difficult to purchase low-cost properties using traditional mortgage financing because the loan size is too small. This affects housing stock of all kinds, including singlefamily, site-built, and manufactured housing, as well as condominiums, which are all sources of affordable single-family housing for low- and moderate-income households. Our nation is facing an affordable housing crisis, where high rents and slow and variable income growth, combined with renewed household formation, have generated a growing and unmet demand for more affordable housing. The supply of low-cost properties could be a part of the solution, but first-time homebuyers and those with limited funds who cannot buy these properties outright with cash need access to housing finance solutions. Given the age of the single-family housing stock in many low-cost markets, affordable finance for purchase and rehabilitation as well as renovation options for families needing reinvestment in their homes to retain value and grow equity should be important components of any solution. Our research indicates that ideas such as the following could create new sources of capital and financing that could improve opportunities for LMI families who wish to purchase, rehabilitate, or refinance with a small-dollar mortgage. They deserve further exploration and experimentation. Review regulations and business practices in the real estate and mortgage finance space that might be creating barriers to lending on low-cost properties. Taking a fresh look at government lending guidelines for the FHA and VA and at opportunities for conventional conforming financing of low-cost single-family properties could bring needed liquidity to LMI borrowers. The FHA and VA supplied surprisingly few of the small loans originated in recent years, and with higher interest rates and lower DTI ratios than larger loans, despite only modestly lower borrower FICO scores. Reviewing mortgage underwriting and lending practices with an eye toward low-cost markets could uncover areas where traditional mortgage lending rules and practices may be putting potential LMI borrowers at a huge disadvantage. Expand the role of local and federal government, the secondary market (the GSEs), and community-based organizations in small-dollar housing finance. Introducing new programs and leveraging the capacity of community-based nonprofit organizations, land banks, state housing finance agencies, credit unions, and community development financial institutions could expand the market for small-dollar mortgage lending. This includes experimenting with new partnerships with public and private groups and exploring new models for bringing liquidity to the small-dollar market through capital markets execution. Leveraging the Federal Housing Finance Agency Duty to Serve rules to help foster and promote more active lending for SMALL- DOLLAR MORTGAGES FOR SINGLE- FAMILY RESIDENTIAL PROPERTIES 25

rural and underserved markets, as well as testing new products or instruments to address these challenges and expand lending to LMI households would be an ideal outcome as Duty to Serve market plans are implemented. The 2018 scorecard for the GSEs requires the enterprises to safely expand access to credit for creditworthy and underserved borrowers and markets (FHFA 2017). As they assess the availability of low-balance loan financing and define approaches to making access to financing more available, new partnerships and product innovations should be considered. Create new consumer-friendly, fairly priced small-dollar mortgage alternatives to traditional mortgages for home purchase, renovation, and refinance. The typical mortgage process might not be the best way to serve and lend to prospective borrowers who are looking to enter homeownership. Mortgage products tend to be long term and subject to primary and secondary market requirements that may be overly burdensome for small loans. Several parts of the mortgage process and the associated fees and costs make it prohibitive for lenders to make money, make it excessively expensive for low-income borrowers who wish to buy, and give little incentive to other players in the real estate market (e.g., real estate agents) to participate. Exploring waivers on appraisals, standardizing loan officer compensation for smaller loans, or streamlining other parts of the mortgage process that would speed up financing and the ability to close would help buyers who finance their purchases with a small mortgage be more competitive with cash buyers. Incentives for sellers and investors of low-cost properties to consider owner-occupant buyers first could help with cash-buyer competition. Expand first look programs that allow first-time homebuyers, low-income borrowers, and minorities in affordable communities the ability to purchase over cash investors through a fund or pool. Establishing and testing programs with large single-family investors and with some home improvement retailers to promote mobility from renting to owning is another avenue to explore. Some single-family rental investors and community nonprofits that own real estate have made significant investments in improving properties and supporting families who wish to become homeowners. More partnerships, collaboration, and cooperation among investorowners, community nonprofits, lenders, and potential borrowers to restore and revitalize lowcost communities could change the trajectory of these properties and convert them from renter occupied to owner occupied. Explore opportunities to improve and expand secured, affordably priced manufactured housing finance, including safer, scalable, market-priced, secured products for chattel lending. Looking at ways to bring manufactured housing finance into a more mainstream and affordably 26 SMALL- DOLLAR MORTGAGES FOR SINGLE- FAMILY RESIDENTIAL PROPERTIES

priced lending structure could increase the production and sale of manufactured housing as an affordable option. This is especially important for resale of manufactured housing a more common transaction as the quality of such housing has improved. Conclusion The limited access to mortgage credit for low-cost properties has led to a growing imbalance in America s housing that affects both demand and supply. Numerous studies have pointed out a growing affordability crisis, for both renters and homeowners, with a widening gap in the affordable housing supply available for families in need. America s housing infrastructure is aging, and economic forces continue to drive home prices up, putting increased pressure on rents and rendering homeownership out of reach for many would-be first-time homebuyers and low-income families. Yet, large swaths of the housing stock provide substantial opportunities for low-cost homeownership. Our analysis shows that sales of low-cost housing are dominated by cash buyers and investors, and borrowers who are creditworthy and able to purchase a low-cost property with a mortgage (in the agency market) typically pay a higher interest rate and put more money down in percentage terms, despite having similar credit scores and lower debt-to-income ratios. Addressing this access-to-credit issue for small mortgages would extend opportunities for affordable homeownership to additional creditworthy borrowers, especially in low-cost markets, which include many rural, urban, and suburban areas. In addition, small refinance and renovation loan options could help owners with low-balance properties improve their homes and assist with needed changes that will improve livability and neighborhood appeal. Building robust tools to support small-dollar mortgage lending could expand the opportunity to turn renters especially traditionally underserved families, such as low- and moderate-income households, first-time homebuyers, and minority households into homeowners. It could also bring more capital for renovation to homeowners and provide much-needed investment in our naturally occurring affordable single-family housing stock, an investment that is needed now so that future generations can benefit. SMALL- DOLLAR MORTGAGES FOR SINGLE- FAMILY RESIDENTIAL PROPERTIES 27

Notes 1 We exclude the limited number of properties sold for less than $10,000 because these transactions are often not arms-length sales. 2 Ellen Seidman and Bing Bai, Where Have All the Small Loans Gone? Urban Wire (blog), Urban Institute, April 18, 2016, https://www.urban.org/urban-wire/where-have-all-small-loans-gone. 3 Alanna McCargo and Sarah Strochak, Mapping the Black Homeownership Gap, Urban Wire (blog), Urban Institute, February 26, 2018, https://www.urban.org/urban-wire/mapping-black-homeownership-gap. 4 Laurie Goodman, Edward Golding, Alanna McCargo, and Bhargavi Ganesh, Manufactured Homes Could Ease the Affordable Housing Crisis. So Why Are So Few Being Made? Urban Wire (blog), Urban Institute, January 29, 2018, https://www.urban.org/urban-wire/manufactured-homes-could-ease-affordable-housing-crisis-so-whyare-so-few-being-made. 5 Small Dollar Credit, Center for Financial Services Innovation, accessed April 3, 2018, https://cfsinnovation.org/research/small-dollar-credit/. 6 See page 4 for a discussion of why we chose $70,000 as the defining price and mortgage size. 7 Property sales records were obtained from the Deeds file of CoreLogic Real Estate data. By comparing mortgages up to $70,000 with properties and home sales up to $70,000, we are assuming zero down payment. Small-dollar mortgage loans generally have high loan-to-value ratios (the median loan-to-value ratio on agency loans in 2017 was 87 percent). Because a $70,000 loan with the standard 3.5 percent Federal Housing Administration (FHA) down payment would support a home purchase price of $72,539, our analysis might understate the difficulty of purchasing a low-cost home with even an FHA mortgage but especially with government-sponsored enterprise financing (see table 2 for credit characteristics of agency loans by channel and loan size). That is, if you consider borrowers who have a down payment and could thus afford a home worth more than $70,000, the small-dollar mortgage shortfall is even more significant. 8 Housing Credit Availability Index, Urban Institute, accessed April 3, 2018, https://www.urban.org/policycenters/housing-finance-policy-center/projects/housing-credit-availability-index. 9 See Laurie Goodman, Jun Zhu, and Bing Bai, Overly Tight Credit Killed 11 Million Mortgages in 2015, Urban Wire (blog), Urban Institute, November 21, 2016, https://www.urban.org/urban-wire/overly-tight-credit-killed- 11-million-mortgages-2015. Calculations updated with 2016 Home Mortgage Disclosure Act data. 10 The property sales data are based on CoreLogic property record data. Our data have good market coverage through 2015. 11 Ellen Seidman and Bing Bai, Where Have All the Small Loans Gone? Urban Wire (blog), Urban Institute, April 18, 2016, https://www.urban.org/urban-wire/where-have-all-small-loans-gone. 12 The counties with missing 2015 sales data accounted for only 1.6 percent of total purchase originations in 2015 based on Home Mortgage Disclosure Act data. 13 A core-based statistical area is a geographic area that includes an urban center of at least 10,000 people as well as one or more adjacent counties. Core-based statistical areas with an urban center with between 10,000 and 50,000 people are micropolitan statistical areas, while those with an urban center with more than 50,000 people are metropolitan statistical areas. 14 The median debt-to-income ratio for mortgages up to $70,000 was 34 percent, and the median note rate was 4.4 percent, in 2017 (table 2). 28 NOTES

15 Of course, potential homeowners have to overcome barriers beyond income. In particular, renters tend to have lower credit scores than owners. Approximately 49 percent of renters have a credit score above 650 (only 33 percent have credit score above 700), compared with 81 percent of homeowners (Li and Goodman 2016). 16 Neighborhood Stabilization Program, US Department of Housing and Urban Development, accessed April 3, 2018, https://www.hudexchange.info/programs/nsp/. 17 Who We Are, Detroit Land Bank Authority, accessed April 3, 2018, https://buildingdetroit.org/overview/. 18 About the Healthy Rowhouse Project, Healthy Rowhouse Project, accessed April 3, 2018, http://healthyrowhouse.org/about/ 19 The core-based statistical area median income was $59,200 for small-dollar mortgage borrowers versus $66,000 for all loans. The census tract median income was $47,632 for new small-loan borrowers versus $65,575 for all new borrowers. NOTES 29

References CFPB (Consumer Financial Protection Bureau). 2014. Manufactured-Housing Consumer Finance in the United States. Washington, DC: CFPB. Engel, Emily, Jason Keller, and Taz George. 2016. Residential Mortgage Lending for Underserved Communities: Recent Innovations. ProfitWise News and Views 1:10 19. FHFA (Federal Housing Finance Agency). 2017. 2018 Scorecard for Fannie Mae, Freddie Mac, and Common Securitization Solutions. Washington, DC: FHFA. Goodman, Laurie, and Christopher Mayer. 2018. Homeownership and the American Dream. Journal of Economic Perspectives 32 (1): 31 58. Goodman, Laurie S., Jun Zhu, and Taz George. 2014. Where Have All the Loans Gone? The Impact of Credit Availability on Mortgage Volume. Journal of Structured Finance 20 (2): 45 53. Li, Wei, and Laurie Goodman. 2016. Comparing Credit Profiles of American Renters and Owners. Washington, DC: Urban Institute. Sarkar, Sohini, and Katharine W. McKee. 2004. The Future of Housing Microfinance in the United States. In Housing Microfinance: A Guide to Practice, edited by Franck Daphnis and Bruce Ferguson, 235 68. Bloomfield, CT: Kumarian Press. 30 REFERENCES

About the Authors Alanna McCargo is codirector of the Housing Finance Policy Center at the Urban Institute, where she focuses on center development and strategy, including the cultivation of innovative research partnerships and initiatives within Urban and with external stakeholders. She is also executive director of Urban s Mortgage Servicing Collaborative. Before joining Urban, McCargo was head of CoreLogic Government Solutions, working with federal and state government agencies, regulators, government-sponsored enterprises, and universities to deliver custom data, analytics, and technology solutions to support housing and policy research. Previously, she held leadership roles with Chase and Fannie Mae, managing mortgage product development, servicing portfolios, policy efforts, and mortgage servicing transformation and alignment efforts. From 2008 to 2011, she worked on behalf of the US Treasury Department on implementing federal housing programs, including the Making Home Affordable loss mitigation and foreclosure prevention initiatives with lenders, servicers, and investors nationwide. McCargo has a BA in communications from the University of Houston and an MBA from the University of Maryland. Bing Bai is a research associate with the Housing Finance Policy Center, where he helps build, manage, and explore data to analyze housing finance trends and related policy issues. Formerly an economic modeling senior at Freddie Mac, Bai conducted research on housing and mortgage markets and developed models to evaluate foreclosure alternatives for nonperforming mortgage loans. He holds a PhD in economics from Clemson University. Taz George is a senior research analyst in the Community Development and Policy Studies Division of the Federal Reserve Bank of Chicago. His research examines a range of community development topics, including housing affordability and access to credit in underserved areas. George holds a bachelor s degree in sociology with honors from Stanford University, and he is an MBA candidate at the University of Chicago Booth School of Business. ABOUT THE AUTHORS 31

Sarah Strochak is a research assistant in the Housing Finance Policy Center. She works with researchers to analyze data, write blog posts, and produce data visualizations for the center s work on access to credit, homeownership, and affordable housing. Strochak received a BA with honors in economics from the University of California, Berkeley, with minors in city and regional planning and geospatial information science and technology. While at Berkeley, she was a student fellow for the University of California Carbon Neutrality Initiative and a research assistant at the Terner Center for Housing Innovation. For her senior honors thesis, she developed a methodology for analyzing mandatory foreclosure mediation laws. 32 ABOUT THE AUTHORS

S TATEMENT OF I NDEPENDENCE The Urban Institute strives to meet the highest standards of integrity and quality in its research and analyses and in the evidence-based policy recommendations offered by its researchers and experts. We believe that operating consistent with the values of independence, rigor, and transparency is essential to maintaining those standards. As an organization, the Urban Institute does not take positions on issues, but it does empower and support its experts in sharing their own evidence-based views and policy recommendations that have been shaped by scholarship. Funders do not determine our research findings or the insights and recommendations of our experts. Urban scholars and experts are expected to be objective and follow the evidence wherever it may lead.

2100 M Street NW Washington, DC 20037 www.urban.org